Round 2 Training DatasetThe data being generated and disseminated is the training data used to construct trojan detection software solutions. This data, generated at NIST, consists of human level AIs trained to perform image classification. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers.

Round1 Holdout DatasetThe data being generated and disseminated is the holdout data used to evaluate trojan detection software solutions. This data, generated at NIST, consists of human level AIs trained to perform a variety of tasks (image classification, natural language processing, etc.). A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers.

This tool calculates metrics for investment analysis documented in NIST Advanced manufacturing Series 200-5. It calculates net present value, internal rate of return, and payback period along with executing sensitivity analysis using Monte Carlo techniques. The tool helps to identify the most economical projects/investments. For instance, it could be used to identify the most economical heating and cooling system or it might be used to rank a set of potential investments.

This is a searchable database of NIST patents

Round 1 Training DatasetThe data being generated and disseminated is the training data used to construct trojan detection software solutions. This data, generated at NIST, consists of human level AIs trained to perform a variety of tasks (image classification, natural language processing, etc.). A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers.

The QIF PMI Report (QPR) software generates a spreadsheet from a QIF (Quality Information Framework) file containing Product and Manufacturing Information (PMI). QIF is a unified XML framework standard for computer-aided quality QIF systems, available free to all implementers. QIF enables the capture, use, and re-use of metrology-related information throughout the Product Lifecycle Management (PLM) and Product Data Management (PDM) domains. QIF was created by the Digital Metrology Standards Consortium.

Round 1 Test DatasetThe data being generated and disseminated is the test data used to evaluate trojan detection software solutions. This data, generated at NIST, consists of human level AIs trained to perform a variety of tasks (image classification, natural language processing, etc.). A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers.

This dataset includes the position data of a two-dimensional gantry system experiment in which the G-code commands for the gantry were transmitted through a wireless communications link. The testbed is composed of four main components related to the operation of the gantry system. These components are the gantry system, the Wi-Fi network, the RF channel emulator, and the supervisory computer. In the experimental study, we run a scenario in which the gantry tool moves sequentially between four positions and has a preset dwell at each of the positions.

This database allows the public to browse and search recent FTZ Board manufacturing approvals.

This tool is used to implement Monte Carlo analysis, which uses probabilistic sensitivity analysis to account for uncertainty. This tool is developed to follow the simulation segment of ASTM E1369. This technique involves a method of model sampling. Specification involves defining which variables are to be simulated, the distribution of each of these variables, and the number of iterations performed. The software then randomly samples from the probabilities for each input variable of interest. Three common distributions that are used include triangular, normal, and uniform.